Experiment dataset supplementary materials for DLfM 2018 submission

<p>This is the experiment dataset supplementary materials for the DLfM 2018 paper submission:</p> <blockquote> <p>An extended jingju solo singing voice dataset and its application on automatic assessment of singing pronunciation and overall quality at phoneme-level</p> </blockquote> <p><strong>Files:&nbsp;</strong></p> <ol> <li>&nbsp;dlfm_experiment_dataset_file_list.ods: recording file names of train, validation and test sets split.</li> <li>&nbsp;dlfm_experiment_dataset_phoneme_numbers.ods: phoneme numbers of each phone class in train, validation and test sets.</li> <li>&nbsp;freesound_extractor_feature_list.ods: freesoundExtractor feature name list used in ANOVA feature analysis.</li> <li>&nbsp;log-mel-scaler-keys-label-encoder.zip: files required for training the embedding model, includes logarithmic features, feature scaler, phoneme dictionary keys and label encoder.</li> <li>anova_analysis_essentia_feature.zip: Essentia&nbsp;freesoundExtractor features of each phoneme for ANOVA analysis.</li> <li>pretrained_embedding_models.zip: classification embedding models pretrained&nbsp;on the below datasets.</li> </ol> <p>The recordings listed in&nbsp;dlfm_experiment_dataset_file_list.ods are taken from a collection of&nbsp;jingju&nbsp;solo singing voice audio datasets, which contains three parts:</p> <ul> <li>Part 1:&nbsp;<a href="https://doi.org/10.5281/zenodo.780559">https://doi.org/10.5281/zenodo.780559</a></li> <li>Part 2:&nbsp;<a href="https://doi.org/10.5281/zenodo.842229">https://doi.org/10.5281/zenodo.842229</a></li> <li>Part 3:&nbsp;<a href="https://doi.org/10.5281/zenodo.1244732">https://doi.org/10.5281/zenodo.1244732</a></li> </ul> <p><strong>Contact information</strong>:</p> <p><em>If you have any question, please contact the authors:</em></p> <p>Rong Gong: Email - rong&lt;dot&gt;gong&lt;at&gt;upf&lt;dot&gt;edu</p>